Declarative modelling for Bayesian inference by shallow embedding

A common problem across science and engineering is that aspects of models have to be estimated from observed data. An instance of this familiar to control engineers is system identification. Bayesian inference is a principled way to estimate parameters: exploiting Bayes~ theorem, an equational proba...

Full description

Bibliographic Details
Main Authors: Nilsson, Henrik, Nielsen, Thomas A.
Format: Conference or Workshop Item
Published: 2014
Subjects:
Online Access:https://eprints.nottingham.ac.uk/32829/
_version_ 1848794499340304384
author Nilsson, Henrik
Nielsen, Thomas A.
author_facet Nilsson, Henrik
Nielsen, Thomas A.
author_sort Nilsson, Henrik
building Nottingham Research Data Repository
collection Online Access
description A common problem across science and engineering is that aspects of models have to be estimated from observed data. An instance of this familiar to control engineers is system identification. Bayesian inference is a principled way to estimate parameters: exploiting Bayes~ theorem, an equational probabilistic model is “inverted”, yielding a probability distribution for the unknown parameters given the observations. This paper presents Ebba, a declarative language for proba- bilistic modelling where models can be used both “forwards” for probabilistic computation and “backwards” for parameter estimation. The novel aspect of Ebba is its implementation: a shallow, arrows-based, embedding. This provides a clear semantical account and ensures that only models that support estimation can be expressed. As arrow-like notions have proved useful in modelling dynamical systems, this might also suggest an approach to an integrated language for modelling dynamical systems and parameter estimation.
first_indexed 2025-11-14T19:17:10Z
format Conference or Workshop Item
id nottingham-32829
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:17:10Z
publishDate 2014
recordtype eprints
repository_type Digital Repository
spelling nottingham-328292020-05-04T16:56:07Z https://eprints.nottingham.ac.uk/32829/ Declarative modelling for Bayesian inference by shallow embedding Nilsson, Henrik Nielsen, Thomas A. A common problem across science and engineering is that aspects of models have to be estimated from observed data. An instance of this familiar to control engineers is system identification. Bayesian inference is a principled way to estimate parameters: exploiting Bayes~ theorem, an equational probabilistic model is “inverted”, yielding a probability distribution for the unknown parameters given the observations. This paper presents Ebba, a declarative language for proba- bilistic modelling where models can be used both “forwards” for probabilistic computation and “backwards” for parameter estimation. The novel aspect of Ebba is its implementation: a shallow, arrows-based, embedding. This provides a clear semantical account and ensures that only models that support estimation can be expressed. As arrow-like notions have proved useful in modelling dynamical systems, this might also suggest an approach to an integrated language for modelling dynamical systems and parameter estimation. 2014-10-10 Conference or Workshop Item PeerReviewed Nilsson, Henrik and Nielsen, Thomas A. (2014) Declarative modelling for Bayesian inference by shallow embedding. In: 6th International Workshop on Equation-Based Object-Oriented Modeling Languages and Tools (EOOLT 2014), 10 Oct 2014, Berlin, Germany. Bayesian inference modelling shallow embedding arrows http://dl.acm.org/citation.cfm?doid=2666202.2666208
spellingShingle Bayesian inference
modelling
shallow embedding
arrows
Nilsson, Henrik
Nielsen, Thomas A.
Declarative modelling for Bayesian inference by shallow embedding
title Declarative modelling for Bayesian inference by shallow embedding
title_full Declarative modelling for Bayesian inference by shallow embedding
title_fullStr Declarative modelling for Bayesian inference by shallow embedding
title_full_unstemmed Declarative modelling for Bayesian inference by shallow embedding
title_short Declarative modelling for Bayesian inference by shallow embedding
title_sort declarative modelling for bayesian inference by shallow embedding
topic Bayesian inference
modelling
shallow embedding
arrows
url https://eprints.nottingham.ac.uk/32829/
https://eprints.nottingham.ac.uk/32829/